1
\$\begingroup\$

Is it because the CPU increases its hertz as necessary?

And when a CPU cycle has to enter a register, does the extra distance in its path measurably affect the speed of that specific cycle? And will it delay the start of the next cycle?

\$\endgroup\$
  • 2
    \$\begingroup\$ It consumes more "battery" because it's doing more. If you sit in a chair all day you need very few calories. If you go running, you need more. \$\endgroup\$ – Majenko Dec 24 '14 at 17:57
5
\$\begingroup\$

There are two effects that influence the amount of power a chip needs. The static power (transistor leakage) and dynamic power. The static power depends on the technology used to build the chip. Therefore even at 0Hz clock the chip draws current. The dynamic power is what you're interested in.

The reason for consuming more power is the number of transistors switching per unit of time. The more transistors switch on or off per second the more power is needed.

Some CPUs indeed throttle their clocks going into different modes. That reduces the number of switches. Others have a capability of turning parts of the CPU off to save energy.

The power heavily depends on which parts of the CPU (die) are used at a given time.

\$\endgroup\$
  • \$\begingroup\$ Voltage also affects static power (and, of course, even within a given process technology device design can influence leakage [even typical values held in SRAM can be exploited to reduce leakage]; minimum-sized transistors tend to switch faster but leak more than larger transistors). Voltage also influences dynamic power (roughly proportional to VVF); the squared voltage term can justify reducing frequency even when such reduces performance (and so increases the static energy used for a task). \$\endgroup\$ – Paul A. Clayton Dec 24 '14 at 21:43
  • \$\begingroup\$ Related question on Super User: Does a CPU consume less power when it is idle? \$\endgroup\$ – Marc.2377 Sep 21 '16 at 13:32
1
\$\begingroup\$

And when a CPU cycle has to enter a register, does the extra distance in its path measurably affect the speed of that specific cycle?

This doesn't make sense, but the answer to the question I think you mean is no; the CPU doesn't know it's own propagation delay. The propagation delay is fixed at design time but affected by manufacturing issues and temperature. If the clock signal arrives too early it malfunctions; this is what happens when a CPU is overclocked too far.

\$\endgroup\$
  • \$\begingroup\$ This, of course, assumes a synchronous design (but "CPU cycle" does not have much meaning in an asynchronous design). It has also been proposed to support variable propagation delay either from voltage variation or input variation, though such designs generally assume a fixed delay and correct for excessive delay (so in that sense the delay is fixed). \$\endgroup\$ – Paul A. Clayton Dec 24 '14 at 21:52

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.